Can Open Models Solve Corporate AI Washing

Can Open Models Solve Corporate AI Washing

More

Descriptions:

The AI Daily Brief examines two stories that together reveal the current state of enterprise AI: Palantir’s record Q2 2026 earnings and Alibaba’s release of Qwen 3.8 Max, one of the most capable open-weight models yet released.

Palantir reported $1.94 billion in quarterly revenue — up 93% year-over-year — with commercial revenue growth accelerating to 149% from 133% in Q1, and net income of $1 billion expanding at a 225% annual pace. CEO Alex Karp used the earnings call and shareholder letter to argue forcefully for what he calls AI sovereignty: the idea that enterprises are at risk of surrendering control over their data, prompts, and organizational intelligence to frontier model providers. Karp explicitly accused AI labs of attempting to “capture the means of production” of their enterprise customers, drawing pointed distinctions between Palantir’s approach and the consumption-driven metrics of large language model providers. In CNBC follow-up comments, he named effective altruism-aligned researchers as central figures in this dynamic.

On the open-source front, Alibaba released Qwen 3.8 Max, a 2.44 trillion parameter model with self-reported benchmarks placing it between GPT-5 and Fable 5 on coding tasks, and state-of-the-art results on OS World Verified — an agentic computer use benchmark. The host argues that enterprise IT buyers, not just developer early adopters, are now actively tracking open-weight releases as part of procurement decisions, citing conversations about fine-tuning strategies and multi-model cost optimization as evidence of rising sophistication. The episode also covers the ongoing Apple versus OpenAI trade secret lawsuit, including OpenAI’s public rebuttal.


📺 Source: The AI Daily Brief: Artificial Intelligence News · Published August 05, 2026
🏷️ Format: Podcast